1. Intelligent Systems and Computational Biology Richard H. Lathrop Dept. of Computer Science rickl@uci.edu Donald Bren Hall 4224 949-824-4021

2. Intelligent Systems and Computational Biology Artificial Intelligence for Biology and Medicine Biology is data-rich and knowledge-hungry AI is well suited to biomedical problems oExamples oMachine learning -- drug discovery oRule-based systems – drug-resistant HIV oHeuristic search -- protein structure prediction oConstraints – design of large synthetic genes oDNA nanotechnology and space-filling DNA tetrahedra oCurrent Project oMachine learning and p53 cancer rescue mutants Goal of talk: The power of information science to influence molecular science and technology

3. “Computers are to Biology as Mathematics is to Physics.” --- Harold Morowitz (spiritual father of BioMatrix, and Intelligent Systems for Molecular Biology Conference) Goal of talk: The power of information science to influence molecular science and technology

4. Biology has become Data Rich Massively Parallel Data Generation Genome-scale sequencing High-throughput drug screening Micro-array “gene chips” Combinatorial chemical synthesis “Shotgun” mutagenesis Directed protein evolution Two-hybrid protocols for protein interaction Half a million biomedical articles per year

5. “Data Rich” Genomic sequence data

6. “Data Rich” Protein 3D structure data Protein Databank Content Growth

7. “Data Rich” Biomedical literature

8. “Data Rich” 10-100K data points per gene chip

9. Characteristics of Biomedical Data Noise!! => need robust analysis methods Little or no theory. => need statistics, probability Multiple scales, tightly linked. => need cross-scale data integration Specialized (“boutique”) databases => need heterogeneous data integration

10. Intelligent Systems are well suited to biology and medicine Robust in the face of inherent complexity Extract trends and regularities from data Provide models for complex processes Cope with uncertainty and ambiguity Content-based retrieval from literature Ontologies for heterogeneous databases Machine learning and data mining Intelligent systems handle complexity with grace

11. Intelligent Systems and Computational Biology Artificial Intelligence for Biology and Medicine Biology is data-rich and knowledge-hungry AI is well suited to biomedical problems oExamples oMachine learning -- drug discovery oRule-based systems – drug-resistant HIV oHeuristic search -- protein structure prediction oConstraints – design of large synthetic genes oDNA nanotechnology and space-filling DNA tetrahedra oCurrent Project oMachine learning and p53 cancer rescue mutants Goal of talk: The power of information science to influence molecular science and technology

12. Drug Discovery Background Cost and time, per drug, start to finish US$500 million to US$1 billion 6 to 12 years High failure rate “Blockbuster” drugs Revenues > US$1 billion/year Profits > US$1 million/day Blood pressure, ulcers, …. A risky business….

14. Positive Examples * Negative Examples

15. Digital 3D Shape Representation

16. The Multiple Instance Problem “Solving the multiple instance problem with axis-parallel rectangles” Dietterich, Lathrop, Lozano-Perez, Artificial Intelligence 89(1997) 31-71

17. “Compass: A shape-based machine learning tool for drug design,” Jain, Dietterich, Lathrop, Chapman, Critchlow, Bauer, Webster, Lozano-Perez, J. Of Computer- Aided Molecular Design, 8(1994) 635-652

18. New Industry: Arris Pharmaceutical Corp. Started a Venture Pharmaceutical Company Apply machine learning to drug discovery Became a Publicly Traded Company (1993) Value of $60 Million, 59 Full-time Employees Merged with Sequana Pharmaceuticals (1998) Merger result became Axys Pharmaceuticals Finally bought by Celera Genomics (2001) Became Celera Therapeutics

19. Intelligent Systems and Computational Biology Artificial Intelligence for Biology and Medicine Biology is data-rich and knowledge-hungry AI is well suited to biomedical problems oExamples oMachine learning -- drug discovery oRule-based systems – drug-resistant HIV oHeuristic search -- protein structure prediction oConstraints – design of large synthetic genes oDNA nanotechnology and space-filling DNA tetrahedra oCurrent Project oMachine learning and p53 cancer rescue mutants Goal of talk: The power of information science to influence molecular science and technology

20. Knowledge-based avoidance of drug-resistant HIV mutants Physician’s advisor for drug-resistant HIV Rules link HIV mutations & drug resistance Rules extracted from literature manually Patient’s HIV is sequenced Rules identify patient-specific resistance Rank approved combination treatments

21. Input/Output Behavior INPUT = HIV SEQUENCES FROM PATIENT: 5 HIV clones (clone = RT + PRO) = 5 RT + 5 PRO (RT = 1,299; PRO = 297) = 7,980 letters of HIV genome OUTPUT = RECOMMENDED TREATMENTS: 12 = 11 approved drugs + 1 humanitarian use Some drugs should not be used together 407 possible approved treatments

22. Example Patient Sequence of HIV Reverse Transcriptase (RT) CCA GTA AAA TTA AAG CCA GGA ATG GAT GGC CCA AAA GTT AAA CAA TGG CCA CCC ATT AGC CCT ATT GAG ACT GTA TTG ACA GAA GAA AAA ATA AAA GCA TTA GTA GAA ATT TGT ACA GAG ATG GAA AAG GAA GGG *AA ATT TCA AAA ATT GGG CCT GAA AAT CCA TAC AAT ACT CCA GTA TTT GCC ATA AAG AAA AAA GAC AGT ACT AAA TGG AGA AAA TTA GTA GAT TTC AGA GAA CTT AAT AAG AGA ACT CAA GAC TTC TGG GAA GTT CAA TTA GGA ATA CCA CAT CCC GCA GGG TAA AAA AAG AAA AAA TCA GTA ACA GTA CTG GAT GTG GGT GAT GCA TAT TTT TCA GTT CCC TTA GAT GAA GAC TTC AGG AAG TAT ACT GCA TTT ACC ATA CCT AGT ATA AAC AAT GAG ACA CCA GGG ATT AGA TAT CAG TAC AAT GTG CTT CCA [CAG] GGA TGG AAA GGA TCA CCA GCA ATA TTC CAA AGT AGC ATG ACA AAA ATC TTA GAG CCT TTT AGA AAA CAA AAT CCA GAC ATA GTT ATC TAT CAA TAC ATG GAT GAT TTG TAT GTA GGA TCT GAC TTA GAA ATA GGG GAG CAT AGA ACA AAA ATA GAG GAG CTG AGA CAA CAT CTG TTG AGG TGG GGA CTT ACC ACA CCA GAC AAA AAA CAT CAG AAA GAA CCT CCA TTC CTT TGG ATG GGT TAT GAA CTC CAT CCT GAT AAA TGG ACA GTA CAG CCT ATA GTG CTG CCA GAA AAA GAC AGC TGG ACT GTC AAT GAC ATA CAG AAG TTA GTG GGG AAA TTG AAT TGG GCA AGT CAG ATT TAC CCA GGG ATT AAA GTA AGG CAA TTA TGT AAA CTC CTT AGA GGA ACC AAA GCA CTA ACA GAA GTA ATA CCA CTA ACA GAA GAA GCA GAG CTA GAA CTG GCA GAA AAC AGA GAG ATT CTA TAA GAA CAA GTA CAT GGA GTG TAT TAT GAC CCA TCA AAA GAC TTA ATA GCA GAA ATA CAG AAG CAG GGG CAA GGC CAA TGG ACA TAT CAA ATT TAT CAA GAG CCA TTT AAA AAT CTG AAA ACA GGA AAA TAT GCA AGA ATG AGG GGT GCC CAC ACT AAT GAT GTA AAA CAA ATA ACA GAG GCA GTG CAA AAA ATA ACC ACA GAA AGC ATA GTA ATA TGG TGA AAG ACT CCT AAA TTT AAA CTG CCC ATA CAA AAG GAA ACA TGG GAA ACA TGG TGG ACA GAG TAT TGG CAA GCC ACC TGG ATT CCT GAG TGG GAG TTT GTT AAT ACC CCT CCC ATA GTG AAA TTA TGG TAC CAG TTA GAG AAA GAA CCC The bracketed codon [CAG] causes strong resistance to AZT.

23. Rules represent knowledge about HIV drug resistance IF THEN [weight] (reference). IF RT codon 151 is ATG THEN do not use AZT, ddI, d4T, or ddc. [weight=1.0] (Iversen et al. 1996) The weight is the degree of resistance, NOT a confidence or probability.

24. “Knowledge-based avoidance of drug- resistant HIV mutants” Lathrop, Steffen, Raphael, Deeds-Rubin, Pazzani, Cimoch, See, Tilles AI Magazine 20(1999)13-25

25. Intelligent Systems and Computational Biology Artificial Intelligence for Biology and Medicine Biology is data-rich and knowledge-hungry AI is well suited to biomedical problems oExamples oMachine learning -- drug discovery oRule-based systems – drug-resistant HIV oHeuristic search -- protein structure prediction oConstraints – design of large synthetic genes oDNA nanotechnology and space-filling DNA tetrahedra oCurrent Project oMachine learning and p53 cancer rescue mutants Goal of talk: The power of information science to influence molecular science and technology

26. Protein structure prediction

27. “Protein Threading”

28. “Global Optimum Protein Threading with Gapped Alignment and Empirical Pair Score Functions” Lathrop and Smith J. Mol. Biol. 255(1996)641-665

29. Multi-queue Branch-and-Bound Originally developed for protein structure prediction by “protein threading” Abstracted to a general-purpose method Have also run on SAT, Traveling Salesman Small molecule conformation search, DNA motif discovery, DNA nanotechnology, synthetic gene design and assembly Difficult applications inspire new techniques. New techniques enable other applications.

30. Optimal Exponent < Proof Exponent X axis = log10(search space size), Y axis = log10(time in seconds) “x” = time to best result, “o” = time to optimal, “*“ = search abandoned Small organic molecule conformation search Protein-DNA binding motif search

31. “A multi-queue branch- and-bound algorithm for anytime optimal search with biological applications” Lathrop, Sazhin, Sun, Steffen, Irani Genome Informatics 12(2001)73-82 GIW’2001, Tokyo, Japan

32. Intelligent Systems and Computational Biology Artificial Intelligence for Biology and Medicine Biology is data-rich and knowledge-hungry AI is well suited to biomedical problems oExamples oMachine learning -- drug discovery oRule-based systems – drug-resistant HIV oHeuristic search -- protein structure prediction oConstraints – design of large synthetic genes oDNA nanotechnology and space-filling DNA tetrahedra oCurrent Project oMachine learning and p53 cancer rescue mutants Goal of talk: The power of information science to influence molecular science and technology

33. Current Application: Synthetic gene design Constraint-based search through sequence space Constraints guarantee correct self-assembly and other desirable properties Underlying domain-specific computation done on a 64-node cluster of 3GHz Xeon dual processors

34. Assembly of Integrase Gene (1640 bp)

35. Problem: Melting Temperatures of correct and incorrect hybridization assemblies overlap solid: correct overlap of oligos dashdot: correct overlap of intermediate fragments dashed: incorrect overlap of small oligos dotted: incorrect overlap of intermediate fragments - CODA Result: Intermediate assemblies contain multiple fragments and error products (smears above)

36. Solution: Fragments designed with non-overlapping melting temps + CODA solid: correct overlap of oligos dashdot: correct overlap of intermediate fragments dashed: incorrect overlap of small oligos dotted: incorrect overlap of intermediate fragments Result: Single products assembled correctly.

37. In vivo Expression of Gag protein in E. coli Confirmed by Western blot and sequencing

38. Intelligent Systems and Computational Biology Artificial Intelligence for Biology and Medicine Biology is data-rich and knowledge-hungry AI is well suited to biomedical problems oExamples oMachine learning -- drug discovery oRule-based systems – drug-resistant HIV oHeuristic search -- protein structure prediction oConstraints – design of large synthetic genes oDNA nanotechnology and space-filling DNA tetrahedra oCurrent Project oMachine learning and p53 cancer rescue mutants Goal of talk: The power of information science to influence molecular science and technology

39. 3D DNA Nanostructures Christopher D. Wassman UC Irvine Dept. of Computer Science

40. Why DNA Nanotechnology DNA has an well understood 3D structure DNA is easily synthesized and manipulated DNA Feature Sizes: 3.6 nm per helical rise, 2 nm helical width Intel Feature Sizes: Current chips, 45nm feature size Research chips, 32nm feature size (Sept, 2008) Bio-Nanotechnology is a emerging field Lots to do, and lots of fun to be had!

41. Tiling 3-Space A familiar concept Building blocks Cubes fill space Cylinders do not Other building blocks are possible We will focus on tetrahedral building blocks, constructed by “folding DNA”

42. Irregular Tetrahedra… Can Tile 3-Space Completely!

43. Full Tetrahedron

44. A Closer Look

45. Atomic Force Microscopy (AFM)

46. Experimental AFM Image

47. Intelligent Systems and Computational Biology Artificial Intelligence for Biology and Medicine Biology is data-rich and knowledge-hungry AI is well suited to biomedical problems oExamples oMachine learning -- drug discovery oRule-based systems – drug-resistant HIV oHeuristic search -- protein structure prediction oConstraints – design of large synthetic genes oDNA nanotechnology and space-filling DNA tetrahedra oCurrent Project oMachine learning and p53 cancer rescue mutants Goal of talk: The power of information science to influence molecular science and technology

48. p53 is a central tumor suppressor protein “The guardian of the genome” Cancer Mutants: About 50% of all human cancers have p53 mutations. p53 core domain bound to DNA Image Generated with UCSF Chimera Cho, Y., Gorina, S., Jeffrey, P.D., Pavletich, N.P. Crystal structure of a p53 tumor suppressor-DNA complex: understanding tumorigenic mutations. Science v265 pp.346-355, 1994 p53 and Human Cancers

49. Consequences of p53 mutations Cho et al., Science 265, 346-355 (1994) Loss of DNA contactDisruption of local structure Denaturation of entire core domain p53 regulates several hundred downstream genes Involved in cell cycle, DNA damage repair, apoptosis

50. Suppressor Mutations Several second-site mutations restore functionality to some p53 cancer mutants in vivo. NC Core domain for DNA bindingTetramerization 102-292 324-355 Transactivation 1-42 175 245 248 249 273 282 C S

51. Will not grow. Will grow. INACTIVE ACTIVE Baroni, T.E., et al., 2004 High-throughput p53 Transcription Assay Initial: Yeast Growth Selection, Sequencing Confirm: Human 1299 Cell-based Luciferase Human p53 consensus (S) = Strong (W) = Weak (N) = Negative Danziger, S.D., et al., 2009Baronio, R., et al., 2010 URA− First measurement Firefly luciferase p53 dependent Second measurement Renilla luciferase p53 independent

52. NC Core domain for DNA bindingTetramerization 102-292 324-355 Restore integrity of p53 core domain Transactivation 1-42 C A Long-held Goal of Anti-cancer Therapy Restore apoptosis (cell death) in tumor cells

53. Theory Find New Cancer Rescue Mutants Knowledge Experiment Active Machine Learning for Biological Discovery

54. Spiral Galaxy M101 http://hubblesite.org/ ~10^9 stars. ~10^9 stars. Known Mutants ~167 stars Known Actives ~1 star Known Mutants: 16,722 Known Actives: 143 Assuming up to 5 mutations in 200 residues How Many Mutants are There?: ~10^11

55. Example M … Example N+4 Example N+3 Example N+2 Example N+1 Unknown Example N … Example 3 Example 2 Example 1 Known Training Set Classifier Train the Classifier Add New Examples To Training Set Choose Examples to Label Computational Active Learning Pick the Best (= Most Informative) Unknown Examples to Label

56. MIP Positive (96-105) MIP Negative (223-232) Expert (114-123) # Strong Rescue 80 (p < 0.008)6 (not significant) # Weak Rescue32 (not significant)7 (not significant) Total # Rescue112 (p < 0.022)13 (not significant) p-Values are two-tailed, comparing Positive to Negative and Expert regions. Danziger, et al. (2009) Novel Single-a.a. Cancer Rescue Mutants No significant differences between the MIP Positive and Expert regions. Both were statistically significantly better than the MIP Negative region. The Positive region rescued for the first time the cancer mutant P152L. No previous single-a.a. rescue mutants in any region.

57. Cancer Mutation Inactive p53 Anti- Cancer Drug += Active p53 Mutations Rescue Cancerous p53 Cancer Mutation Inactive p53 Wild Type Active p53 Cancer+Rescue Mutations Active p53 Cancer Cancer Ultimate Goal

58. The long road to a future anti-cancer drug IIIIIIIVV NC C S drug Peter Kaiser Rommie Amaro Dick Chamberlin Melanie Cocco Hudel Luecke Wes Hatfield Chris Wassman Roberta Baronio Ozlem Demir Faezeh Salehi Edwin Vargas Da-Wei Lin

59. Intelligent Systems and Computational Biology Artificial Intelligence for Biology and Medicine Biology is data-rich and knowledge-hungry AI is well suited to biomedical problems oExamples oMachine learning -- drug discovery oRule-based systems – drug-resistant HIV oHeuristic search -- protein structure prediction oConstraints – design of large synthetic genes oDNA nanotechnology and space-filling DNA tetrahedra oCurrent Project oMachine learning and p53 cancer rescue mutants Goal of talk: The power of information science to influence molecular science and technology

61. NC 175 245 248 273 282 249 Core domain for DNA bindingTetramerizationTransactivation IARC TP53 Mutation Database (R7, Sep. 2002): 17,689 somatic mutations 12,631 (71%) missense mutations affecting codons 100-300 974 different amino acid changes in core domain The spectrum of p53 mutations

63. Classifier & Feature Overview p53 Wild type Crystal p53 Mutant Structure Amber™ Homology Modeler Mutant Residue (s) Features Classifier Mutant Functionality

64. Multi-dimensional Features 2D Surface Map 3D Structural Changes 4D Unfolding Trajectory Danziger et al., Functional Census of Mutation Sequence Spaces, IEEE/ACM Trans Comput Biol Bioinform (2006)

65. Cancer Mutant Rescued Cancer Mutant PREDICT

66. Simulated AFM Image X axis (nanometers) Y axis (nanometers)

67. Experimental AFM Image Y axis (nanometers) X axis (nanometers)